What's new
What's new

Mazak QT 10N (1986) Which bearing numbers to use in the spindle?

cwtoyota

Cast Iron
Joined
Feb 11, 2010
Location
Washington State
I picked up this lathe for a low price knowing it was a "pig in a poke". It needs a spindle rebuild.
I will be performing the work myself, but I need to find out a couple of bearing numbers in order to proceed.


I have the spindle out and disassembled on the bench. The tapers and bearing surfaces look good. There's a little discoloration at the working end of the spindle due to heat from the doomed bearings, but nothing seized or spun.

Only two of the four bearings were still in there and all of the seals were gone. I assume someone previously thought they would fix the spindle and when faced with the cost and complexity of the job they decided it wasn't worth it for an old machine.

The two bearings I do have are the NN3016K CC1P5 at the far end of the spindle and the big NN3020 CC9P4 (missing some rollers) at the business end.

I've done some reading of other posts here and determined that I will need to know the correct contact angle and pre-load to choose the correct 7020 P4 bearings as a set. Any input on this is appreciated.

I don't know if the seals are Mazak proprietary parts, or if I can get them through SKF, etc. If anyone can help me with seal numbers that would be great too.

Also, isn't it nice of Mazak to drill and tap holes in everything for push-bolts. Everything comes out nice and smooth without pullers, presses, heat or drifts and hammers.

IMG_1499.jpgIMG_1497.jpgIMG_1495.jpgIMG_1498.jpg
 
couldn't find drawing for a qt10 but found a qt12 it shows the labyrinth seals as "collar" no description just part number. the numbers wont work for you as it does use different bearing from the ones you listed.
 
Thanks for looking.
I have the steel labyrinth collar that goes on the 100mm bearing bore, between the front bearing and the flange on the spindle.

What I seem to be missing is some kind of replaceable (rubber?) element that fits in an ID groove on the outermost shroud of the working end of the spindle. There is a similar groove inside the rear bearing housing the small end of the spindle. These grooves are about 3/8" (10mm ?) wide and equally deep (radial depth). There are corresponding sawtooth grooves in the rotating elements that suggest some kind of rubber strip or ring goes in there as a packing.

Any chance I can get my hands on that QT12 drawing?
 
I heard the bull roarer. Crocodile Dundee II (1988) clip 2 on ASO - Australia's audio and visual heritage online

I seriously doubt you are missing a rubber lip seal there. Seals produce heat, worst possible place to produce it. The spindle grows, you will have labyrinth seals, male and female grooves and lands that interconnect each other with a minimum clearance.

I've been inside a lot of Mazak Spindle's and have never found a true seal on a spindle. Seals do exist, but normally reserved to high end German machines. Bohringer do that, but they run them on Ceramic Metal Sprayed journals.

The bearings you are missing would have had boundary dimensions of a 7020. If they are tucked behind a NN3020K. 100 x 150 x 24mm

You have the NN 3016 & NN 3020 under your nose. I think we could agree they have absolutely no capacity for axial loads.

The angular contacts you seek, are often in a different section to the regular angular contacts of a catalog. It will be in a whole separate section from spindle bearings, called Angular Contact Thrust ball bearings.

As the only thing that can take axial / thrust load. They never came at less than 30 degrees. 40 was common, and they went as high as 60 degrees.

My prediction of your missing bearings that no one has ever seen is that they were:
NSK 100BA10XTYDBLP4A

Being a "100" bore, "BA" 30 deg, (BT = 40deg) "10X" mates with NN30 series, "TY" = molded cage, "DB" = back to back. "L" = light preload, "P4A"= P4 / ABEC 7 accuracy, but they have kept the pair (2) with in half the accuracy bore grade.

Limiting speed factor of that spindle will always be the NN3020K, Its the biggest element on that axis . Double row roller element compared to the single row angular contact. I engineer contact angle and pre-load up until the speed of the roller bearings. Long story short, I can stiffen a spindle up.

Your NN 3016K & NN 3020K. You cant just buy a new set and slap them in there. They are a black art, your lucky day, I know a bit about them.

Regards Phil.
 
I am not sure if this is heresy ...
But very cheap industrial zero-contact airtight lab seals exist.

Nilos rings.
== 20$ each, for 7010 or so.

Google it.
Run-in 24 hours, they lap in place, and last forever, zero contact.

I know it is not stock, but ...
 
no rubber at the front end is gospel on every Mazak I've seen.
we still need to hope he is able to bring you to speed on the whole process of measuring the spacer gap, calculating the spacer grind based on radial clearance and all the rest of the black art of the spindle deal.
you have the Masters ear grasshopper, pay attention.
 
I am not sure if this is heresy ...
But very cheap industrial zero-contact airtight lab seals exist.

Nilos rings.
== 20$ each, for 7010 or so.

Google it.
Run-in 24 hours, they lap in place, and last forever, zero contact.

I know it is not stock, but ...

there is no place to put a seal in the front end of a Mazak.
the groove of which he speaks is what creates the air pressure that makes a laberant sealing system works. if you stuff rubber in there it fails the whole system.
lathe spindles are not the place to experiment with diy bullshit. the bearings Mac is talking about are $5k a set. you screw with that setup without the wisdom of decades of building spindles and your gonna end up with shit.

do what Phil says.
 
you know the girls at my place took out a qt10 hanging a bar out the back.
better check that tube, .050 run output on the back end could have been the show stopper on the guy that tore it apart
 
Thanks EVERYONE for your input!

I understand the concept of labyrinth seals, but this is my first spindle and my first practical experience with these seals. I saw a groove and ASSUMED there must be something meant to go in it. Now that I look at it again, there are large drain holes at the bottom of each groove. The grooves are also mildly dirty as if they have never had a seal in them.

On to the thrust bearings, I understand that they are factory ground to create the design pre-load when placed back to back and clamped to the correct torque.

So given that I can find a matched set of new NSK 100BA10XTYDBLP4A and a set of rollers, I should be in business.

I know just enough about the rollers to get myself into trouble. I have CC9 on the NN3020 bearing and CC1 on the NN3016. Does that tell me the radial clearance in micron when the bearing was new? Should I be looking for bearings with the same, or more clearance to avoid shims?

I understand that we have a 1:12 taper to expand the inner race and eliminate the radial clearance. Knowing how far to go onto the taper sounds like it's done by feel? Also, what is the best way to measure the spacer gap?

I will locate a set of those thrust bearings and purchase them. I'll hold off on the roller bearings until I'm up to speed on which numbers I should use and the procedure for measuring them.

Thanks again, for sharing your knowledge with me here.

-Chris
 
you know the girls at my place took out a qt10 hanging a bar out the back.
better check that tube, .050 run output on the back end could have been the show stopper on the guy that tore it apart

That's a good suggestion. I will have it between some v-blocks today and see what kind of run-out I have at the nose and a few other ground locations.

I suppose I can use the journal radii to calculate my v-block heights:

xx16 bearing x 5mm = 80mm I.D. xx20 bearing x 5mm = 100mm I.D.

Radius 50mm - Radius 40mm = 10mm Delta-R. 10mm = .3937"

So I need .3937" of gauge blocks under the v-block on the small end of the spindle.
 
Thats not what I said.
Please read my post.

Nilos rings are industrial, well proven solutions with ZERO CONTACT as I said.
No rubber.
Zero friction.

there is no place to put a seal in the front end of a Mazak.
the groove of which he speaks is what creates the air pressure that makes a laberant sealing system works. if you stuff rubber in there it fails the whole system.
lathe spindles are not the place to experiment with diy bullshit. the bearings Mac is talking about are $5k a set. you screw with that setup without the wisdom of decades of building spindles and your gonna end up with shit.

do what Phil says.
 
knowing how far to go onto the taper is a distinct formula I hope Phil can help you with.
there is then a spacer that is ground to maintain that exact amount of engagement.

time for the bearing whisperers next post, take it away oz please
 
I read what you said
you suggest other than OEM seal

being a proven industrial device is of little significance here.

thermite is a proven industrial tool. but I don't use it for welding jewelry.

what is of significance is knowing how to build a high speed industrial spindle that will not take water while spinning at 4000 rpm with a coolant line blowing right at the crack.

I don't know how to do this, I suggest your knowledge on this subject may be less than Mazaks.

so all verbage aside

you say change Mazaks system
I say don't

all the rest us a bunch of internet chit chat.


Hey guys, no need to have the dispute. I appreciate the input from both of you guys. I am intent on rebuilding this old spindle to original Mazak spec as best I can. They were solid machines as evidenced by the fact that this one was up and running at least through 2013 (27+ years).

Hanermo, regarding the Nilos rings, I like hearing about clever parts like those because I'll store that info away somewhere in the back of my brain. In 6 months or maybe 6 years I'll have an application in need of that part.
 
the qt12 drawing had these bearings

NSK NN3020K MBCC9P4U17
NSK 100BA10XDBELP4A

Rear

NSK NN3018MBKRCCP5

Rear is different but maybe you can you check this middle bearing to see if works.
 








 
Back
Top